next up previous print clean
Next: Smoothing and damping, accuracy Up: Data examples Previous: Marine 3-D shot interpolation

Interpolating crossline in the 3-D marine case

For some applications, it may make sense to refine the crossline receiver sampling. Prestack datuming, for instance, can be hampered in 3-D because the relatively sparse crossline sampling requires much stronger antialiasing on energy moving in the crossline direction than on energy moving in the inline direction. The SRME demultiple method requires many more receivers than are recorded, in order to model multiples that bounce at surface locations between streamers. In both cases, there are other issues as well. In particular, sources tend to be placed nearly at the center of the crossline range of the receiver spread, and should be extrapolated outward in the crossline direction, for reciprocity's sake. That should probably be done with some other algorithm, because the very narrow dip range that should exist between the two nearly central, only slightly separated sources is not likely to have information necessary to model sources far away. Put another way, PEFs can always interpolate, but can only extrapolate along straight lines.

In the 3-D field data example above, there are only three streamers, with a narrow aperture, so it does not seem especially interesting to interpolate between them. Energy moving in the crossline direction probably becomes more important when there is a larger crossline aperture, as is the case with newer seismic acquisition boats which may tow a dozen or more cables. Unfortunately, no data from those boats was available. Figures m33in and m33out show a shot gather from the SEG-AEG 3-D salt model synthetic, as modeled in the case of Figure m33in and after interpolating streamers between the original streamers in the case of Figure m33out. In this case, 20 or so shot gathers were used in the test. The interpolated data in this case was not modeled, so there are no difference panels, but by inspection the interpolated data seems reasonable. The front face of the two figures do not show exactly the same inline. Instead, Figure m33in shows a particular streamer, and Figure m33out shows a streamer that was interpolated next to it.

 
m33in
m33in
Figure 14
Closeup on a marine 3-D shot gather. Input to the crossline interpolation.
view burn build edit restore

 
m33out
m33out
Figure 15
Closeup on a marine 3-D shot gather. Output of the crossline interpolation.
view burn build edit restore


next up previous print clean
Next: Smoothing and damping, accuracy Up: Data examples Previous: Marine 3-D shot interpolation
Stanford Exploration Project
1/18/2001